This five year research demonstration project proposal is designed specifically to contain the spread of AIDS among intravenous drug users (IVDU) in Los Angeles County. The project will be conducted by the UCLA Drug Abuse Research Group (DARG) in collaboration with Bay Area Addiction Research and Treatment (BAART). Three subsamples of narcotics IVDU shown by recent local seroprevalence surveys to be most at risk for transmission of, or infection by, HIV will be recruited into two subsidized methadone maintenance (MM) programs. For one program, an enhanced MM protocol characterized by contingency management principles and provision of ancillary services will be developed and implemented. A case management system will be designed for the enhanced MM clinic to provide each individual client with the optimal combination of conditions shown to be effective in the clinical trials literature for increasing treatment retention, decreasing IV and other drug use (particularly cocaine), reducing unprotected sexual activities, and reducing relapse after discharge; therefore achieving the ultimate goal of effectively reducing the spread of AIDS among IVDU, and subsequently throughout the general population. Random assignment of clients to the two programs will allow the enhanced program to be evaluated against a regular subsidized MM program so that treatment effects can be unambiguously determined. Evaluation of the programs will occur on major risk reduction criteria, as well as other behavioral domains, on an ongoing basis to provide constant and immediate feedback to the enhanced program for improvement. Secondary aims of the project are to compare the effectiveness of both MM programs to typical Los Angeles County "fee-for-services" programs, as well as to contrast the addiction careers of the proposed subsidized samples against similar data contained in the UCLA data base of 1,800 addiction careers. These comparisons enlarge the scope of the project to a better understanding of the nature of narcotics addiction careers and the effectiveness of various types of treatment across historical periods. Analysis techniques include various multivariate statistical methods such as MANOVA to determine the outcome differences between the two subsidized programs and possible causes and correlates of these differences: repeated measures ANOVA or MANOVA to determine the temporal pattern of changes related to treatment statuses of types; multiple regression analyses to determine if success in the enhanced program is predicted by the same factors as predict success in other kinds of treatment programs; survivorship and hazard rates models to identify risk factors for premature discharge and relapse; and Markov model analyses to determine whether different treatment provisions change the transition probabilities from drug use to abstinence and to relapse. Findings will be disseminated through professional meetings and newsletters, various workshops and training programs, scientific publication and postings to the various nationwide computer bulletin boards maintained by NIDA and NIJ.